PDH locking is much more reliable due to improved alignment.

Its UGF is about 9 kHz with the current setting, but we can set it higher (up to 13 kHz) or lower without touching PLL.

There's some high frequency (about 250kHz) noise in the PDH error signal, which is likely from the PLL.

PLL UGF is also about 9 kHz, which is limited by the PZT resonances (f>100kHz). We'll have to modify the CM board.

We tried to use green WFSs. First we found that there was no bias voltage source so we put an old iLIGO bias unit.  Then we found that the WFS DC interface was not working in Beckhoff world and centering the WFS was hopeless. No PDH DC in Beckhoff, either. We're working on these.

Suspension damping got better than when we started, but the DC level on the PDH diode, depending on the alignment, still fluctuates due to clipping in/on the Faraday by as much as 50%. Better PDH lock means that the cavity wouldn't lose lock with this, but it's a problem nevertheless. For example it's difficult to assess how good or bad the matching is quantitatively. In-lock to out-of-lock ratio of the maximum DC level of the PDH diode was somewhat better than 50%.  Better SUS damping AND larger aperture Faraday should help.

Alexa and Patrick are making some progress and most of slow things in Beckhoff world are being passed to EPICS world already (a notable exception is CM board). Not everything has been tested, though.